On 2/29/24 14:29, Srujana Challa wrote:
This patch series makes Virtio-user works in IOVA as PA mode for vDPA backend. First patch fixes the issue when having buffer IOVA address in control queue descriptors. Second and third patches helps to share descriptor IOVA address, to the vhost backend. And also disables the use_va flag for VDPA backend type. v1->v2: - Split single patch into three patches. Srujana Challa (3): net/virtio_user: avoid cq descriptor buffer address accessing net/virtio: store desc IOVA address in vring data structure net/virtio_user: support sharing vq descriptor IOVA to the backend drivers/net/virtio/virtio_ring.h | 12 ++- .../net/virtio/virtio_user/virtio_user_dev.c | 94 ++++++++++--------- drivers/net/virtio/virtio_user_ethdev.c | 10 +- drivers/net/virtio/virtqueue.c | 4 +- 4 files changed, 69 insertions(+), 51 deletions(-)
I think we need a way to be able to perform the address translation from iova to va between Virtio and Virtio-user layers. Does that sound reasonable? Thanks, Maxime
